Search Images Maps Play YouTube News Gmail Drive More »
Sign in
Screen reader users: click this link for accessible mode. Accessible mode has the same essential features but works better with your reader.

Patents

  1. Advanced Patent Search
Publication numberUS2839692 A
Publication typeGrant
Publication dateJun 17, 1958
Filing dateJul 31, 1956
Priority dateJul 31, 1956
Publication numberUS 2839692 A, US 2839692A, US-A-2839692, US2839692 A, US2839692A
InventorsKirchmayer Leon K
Original AssigneeGen Electric
Export CitationBiBTeX, EndNote, RefMan
External Links: USPTO, USPTO Assignment, Espacenet
Electric power control system
US 2839692 A
Abstract  available in
Images(9)
Previous page
Next page
Claims  available in
Description  (OCR text may contain errors)

June 17, 1958 1 K. KIRCHMAYER 2,839,692

ELECTRIC POWER CONTROL SYSTEM Filed July 32|..l 1955 9 Sheets-Sheet 1 June 1,7, 1958 L. K. KIRCHMAYER 2,839,692

ELECTRIC POWER CONTROL SYSTEM Filed July 51, 195e 9 sheets-sheet 2 lnvenror z Leon K.Kirchmc1yer by M vW7: Z;

His AHorney AMPUHER GENERATOR 75 OUTPUT s\6NAL BALANCTNO AMPUFIER.

GENERATOR OUTPUT SIGNAL GENERATOR FUNCTION GENERATOR Fig. 2

m. 1| m m m C6 "O M CG w n Mm 6 8 w Mm O ONT 7 u O ONT G Rwm G hmmm 6 WWU n KWU 7 0 M L0 8 w H P m8 EP 6 7 u 6 6 r 6 l :Vs: 51| .RJ 5 mm 6 6 mm 6. PM l S I M P0 S M D 8 m 0mm Am T 6 M0. T WD. U LP. U MP lllll lliwp 6 u m HN Tm 45 mum SF. 6 5E 6 r NAE YN 5 YN onu 8| L SU 7 8m NUC S /A 7 o, /3 @5 C 6 \w o OMT@ 2 .l 3 G 5 5 5 4 4 m 5 T C S S S mm E .ML 5 am M 4 YN 7 2 SU RvA 6 8 O A 7 um 4 4 A. 5 4 4 3 m 4/ w m u m N w l. YR 2 R www mm. VF TN .Il 1|. Sill RRl Ul RL E RRQ mm www WN AUE RU U CCR FM A June 17, 1958 L. K. KIRCHMAYER 2,839,692

ELECTRIC POWER CONTROL SYSTEM File@ July 31, 1956 9 Sheets-Sheet 3 pr) Lxg lnvenior t l Leon K.Kirchmo er His Arforney June 17, 1958 L. K. KIRCHMAYER 2,839,692

ELECTRIC POWER CONTROL SYSTEM Filed July 51. 1956 l 9 Sheetsl-Sheet 4 lnvemor Leon K.Kirchmoyer His AHorney June 17, 1958 L. K. KIRCHMAYr-:R 2,839,692

ELECTRIC POWER CONTROL SYSTEM 9 sheets-shea 5 Filed July 3l, 1956 Figa Invenfor Leon K. Kirchmoyer by da... )y ma His Aforney June 17, 1958 L. k. KIRCHMAYER 2,839,692

ELECTRIC POWER CONTROL SYSTEM Filed July' s1, 1956 9 Sheets-Sheet 6 lnvenior Leon K. Kirchmoyer His A'fiorney June 17, 1958 L. K. KlRcHMAYl-:R

ELECTRIC POWER CONTROL SYSTEM Filed July 3l. 1956 9 Sheets-Sheet 7 d3 325 A :3Q- 2F11 1 a? 37 38 FROM TIE LINE CARQWR FREQUENCY FRE UENCY STANDARD, TELEMETERS onf/imm .Tggnm i37| 33| Momrvws l 326 330 c| Rcurr |372 375 ssnvo i M Ampunen l aALANcmG l n Ampuruen 327 373 374 v il BALANCING AMPurn 376 348 Y TIE y.) Loss LINE V n4-- rAcroR AND COMPUTER PLAN-r T *Lomme F" 'll i '-2 l BALANclNs PRoPomlouAL D MPLFER /375 sYsTE 347\ conmouzn M unam- 372i l z I I u l CARRIER S "lY sama 346 AMPunER 35 F|g.lO

368 Invemor: /lll Leon K. Kirchmoyer 367 t His AHorney June 17, 1958 L. K. KIRCHMAYER 2,839,692

ELECTRIC POWER CONTROL SYSTEM Invenfor z Leon K. Kirchmoyer June 17, 1958 L. K. KIRCHMAYER ELECTRIC POWER CONTROL SYSTEM Filed JulyA si. 195e 9 Sheets-Sheet 9 United States Patent ELECTRC PGWER CONTRL SYSTEM Leon K. Kirchmayer, Scotia, N. Y., assigner to General Electric Company, a corporation of New York Application July 3,1., 1956, Serial No. 601,298

51 Claims. (Cl. 307-57) This invention relates to electric power control systems,

, and more particularly to apparatus for controlling most economically, in response to changes in load and fre` quency, the output of a power system which comprises a plurality of interconnected generators and generating stations, which may be connected to other power systems by one or more tie lines. The present invention constitutes an improvement on the power control system which is the subject of U. S. Patent 2,824,240, granted February 18, 1958, on application Serial No. 395,022, tiled November 30, 1953, by E. E. Lynch and l, J. Larew and assigned to the assignee of the present invention.

it is common practice for neighboring power systems to be interconnected by one or more tie lines over which an interchange of power is made according to preselected schedules. It is necessary, of course, that the tie line interchange be held at the previously scheduled values, and the frequency of the generation also held at its desired value.

Generally, each power system includes a plurality of generating stations, each of which contains a plurality of generators. It is known that if a plot is made of fuel input as a function of output power for a generator, the resulting plot, which is known as a fuel-input curve, is usually a curve and not a straight line and it varies from generator to generator. From this curve, another plot may be made of the slope of the fuel-input curve versus power output, which is known as an incremental fuel-rate curve. The incremental fuel rate may be converted to incremental cost by multiplying the incremental fuel rate by the fuel cost. lt is known that, where all generator units are interconnected to supply the same system and transmission losses are ignored, maximum economy is obtained when all generator units are operated at the same incremental fuel cost.

in addition, the generators or generating stations having various incremental fuel costs are generally located at different distances from the load center and so there are various transmission losses which must be taken into account. not it is cheaper to send low-cost power over a long distance than to send high-cost power over a short distance. The transmission losses which any given station must suffer in transmitting its generated power to a distribution point are generally accounted for in computations involving power distribution by a penalty factor assigned to the transmitting station. Therefore, a control system for controlling the power output of these generating stations preferably should embody means for automatic ad justment of the output to permit loading in accordance with incremental fuel costs and with the penalty factors; that is, the output of each generator or station should be automatically adjustable for maximum economy of delivered power.

lt is a primary object of the present invention to provide apparatus for automatically controlling the generation or output of a plurality of generators and generating stations For example, it must be determined whether or' 2,839,692 Patented June 17, 1958 ICC that is capable of loading for maximum economy, holding tie line power interchange to previously arranged schedules automatically, and, simultaneously, holding the system frequency at a predetermined value.

Another object is to provide a load-frequency control system wherein individual signals representing changes in incremental cost of generated power of each generating station may be sent to corresponding stations to cause the power output of the generators therein to vary so that the incremental cost of delivered power is the same for all generating stations.

Another object is to provide a control system wherein a change in power output is made quickly in response to a change in area requirement, and the output of the various generating stations is then reapportioned at a slower rate in accordance with economic loading considerations.

Another object of the invention is to provide a power control system which is capable of effectively controlling the generation of a local power area when the local area is not interconnected with a remote area, or, if a local and remote area are interconnected, is capable of holding the power interchange at a prescheduled value without regard for system frequency, or the system frequency at its predetermined value without regard for power interchange.

A further object of the invention is to provide a system which is as simple and inexpensive in rst cost and maintenance as is consistent with performance of the required functions.

The basic operation of a tie line load-frequency control system can be shown by considering two areas, A and B, interconnected by a tie line and both having their own generation and loads. Economic and other considerations would dictate the most desirable load to send over the tie line, and this could be redetermined as often as required. Generally, the power interchange is rescheduled daily to provide loading for maximum economy and is readjusted at various times during the day of this schedule. In addition to holding the tie line load to the prescheduled value, it is essential that the frequency of the generation in both areas be held at the desired value, which in present day operations is usually 60 cycles per second.

It is known that tie line load and frequency can both be held to the desired values by proper manipulation of the generator outputs at A and at B, in spite of load changes at either A or B. For example, assume that the condition exists wherein the loads at A and B and the scheduled tie line load are exactly supplied. Then, if a reduction in load occurs in area A, it results in an excess of generation at A over the required, which tends to increase both the frequency of generation and the tie line power ow from A to B. Similarly, an increase in load at A results in decreased frequency and decreased tie line power flow from A to B. However, when the load at B decreases, the frequency increases, but the tie line load from A to B decreases. Similarly, when the load at B increases, the frequency decreases, and the tie line load from A to B increases. Therefore, it is apparent that an increase in both frequency and the tie line load from A to B or a decrease in both frequency and tie line load from A to B indicates a load change in area A. Similarly, when the frequency and tie line load change in opposite directions for area A, it indicates a load change in area B. Therefore, by knowing both the change in frequency and change in tie line load, it can be determined whether the load has changed in area A or area B, and the proper adjustments can be made in the generation for these two areas. These facts indicate that for the tie line loadfrequency control, either automatic or manual, at least two primary detectors are required, one which detects frequency and one which detects tie vline load` it is apparent that to restore the system to predetermined or scheduled conditions after the load changes in both areas, corrective action is required `at 'both areas A and B, and if area A properly increases or decreases its generation exactly to supply increases or decreases in load at A, area B will be `required to take care of only its own load changes by corresponding generation changes at B. The system of the present invention is limited to corrective action in only one such area; that is, a control system is required in area A to take care of changes in load and frequency in area A, and another system is required in area B to take care of changes in load and frequency in area B. Thus, a control initiating signal which is a function of frequency divergence and tie line load divergence is required -in order to accomplish this desired control automatically, with the system comprising the present invention.

A load-frequency power control system constructed in accordance with the invention comprises essentially two parts, detector means and control signal producing means located at a central control station, and control signal responsive means located at each of the generating stations comprising the controlled power network. The detector means function to produce a signal Af proportional to the deviation of the actual system frequency from a standard frequency and to producev a signal Aw proportional to the deviation of the actual tie line load from its scheduled value. The control signal producing means combines the Af and Aw signals into a control signal which maybe considered as representing either area power requirement or a change in incremental cost of delivered power.

In accordance with the teachings of the invention, the control signal producing means located at the central control station also embodies means for modifying or varying the control signal for each generating station to cause all stations to operate at equal incremental costs of delivered power at the load center, which is the condition that must be fulfilled for the most economical operation. The control signals may be transmitted to the generating stations by conventional means.

Each generating station embodies control signal responsive means which utilize the control signal to vary the power outputs ofthe various generators within rthe station. The signal responsive means includes means for selecting all or a fractional part of the control signal transmitted thereto to be supplied to individual controls for each generator for varying its generation at a relatively fast rate, and means for each generator acting at a relatively slow rate for converting the control signal to a signal representing desired power output for that particular generator. The generation of each generator is then varied in yaccordance with the corresponding desired power output signal.

The frequency deviation signal Af and the load deviation signal Aw are combined in the detector means at the central control station in such a manner that if the deviations are due to a change in load in area B, nf and nw tend to cancel each other and no change is made in generation in the stations of area A. If the change in load which causes the change in tie line power interchange and the change in system frequency has occurred in area A, which is taken to be the area controlled by the system of the invention, a corrective signal is provided to vary the generation of the system until the condition is corrected.

For a better understanding of the invention, together with further objects and advantages thereof, reference is made to the following description taken in conjunction with the accompanying drawings, in which:

Fig. 1 is a schematic block diagram of one embodiment of detector means and control signal .producing means located at a central control station;

Fig. 2 is a schematic block diagram of one embodiment of the control signal responsive means located at each generating station; i

Fig. 3 is a schematic diagram of a tie line load controller shown in block form in Fig. 1;

Fig. 4 and 4a are schematic diagrams of a servo amplilier that may be embodied in the apparatus shown in Figs. 1 and 2;

Fig. 5 is a schematic diagram of a function generator shown in block form in Figs. 1 and 2;

Figs. 6 and 7 are graphs useful in understanding the operation of the function generator of Fig. 5;

Fig. 8 is a schematic diagram of a balancing amplifier Shown in block form in Figs. 1 and 2;

Fig. 9 is a schematic diagram of an electronic positionunit shown in block form in Fig. 2;

Fig. 10 is a schematic block diagram of a modification of the detector means and control signal producing means installed at a central control station;

Fig. 1l is a schematic diagram of the modifying circuit shown in block form in Fig. 10;

Fig. 12 is a schematic diagram of the proportional controller shown in block form in Fig. l0; and

Fig. 13 is a schematic block diagram of another modification of the central control station detector means and control signal producing means.

The power control system of this invention may be used to control the generation of a local power area, which is interconnected with one or more remote power areas by tie line means, in response to changes in the tie line load and system frequency. Also, the control system may be made responsive to tie line load changes only or to system frequency changes only. However, for purposes of explanation, the more involved system, which is responsive to both t-ie line load and system frequency changes, is illustrated and will be described hereafter.

The tie line load-frequency control system of the invention embodies two primary detectors, one for detecting deviation of the system frequency from a desired value, and one for detecting the tie line load deviation-from its prescheduled value, both forming part of the central control station equipment. Referring now to the embodiment of that equipment shown in Fig. 1, the -means for detecting the deviation of the t-ie line load from the prescheduled value comprises a tie line load controller 2t) and a conventional summing amplifier 21 such as is well known in the art. The input signals to the summing amplier 2i are supplied from conventional telemeter receivers or the like (not shown), such as are available commercially, which provide signals proportional to the actual loads on the tie line means interconnecting the local area with one or more remote areas. The summing amplifier 21. sums the various load signals and provides to the tic line load controller 20 a signal wa proportional to the actual tie line load. Of course, if the system utilizes only a single tie line, the amplifier 21 may be omitted, and the telemetered load signal supplied directly to the tie line load controller 20.

' The tie line load controller 2t) is essentially a servomechanisrn, and any one of numerous known devices may be so employed. One servomechanism, which is known to be suitable for this use, is shown and described in U. S. Patent 2,753,565, granted .luly 3, 1956 on copending application Serial No. 395,117, tiled November 30, 1953, by I. l. Larew and K. N. Burnett, and assigned to the same assignee as the present invention. A sche matic diagram of that device is shown in Fig. 3 and will be described hereafter. However, the present invention is not limited to the use of any one particular device for this purpose. Briefly, the tie line load controller 20 compares the signal wa received from the summing ampliiier 21 with another signal ws, which is produced within the tie line load controller and whose value may be made proportional to the prescheduled value of tie line load, and produces rotation of an output shaft 22 at a speed proportional to the amplitude difference Mw assess@ between the two signals and in a direction determined by the polarity of the difference; k is a proportioning constant whose purpose will be hereafter described.

The output shaft 22 is connected to one input of a conventional mechanical differential 23, whose other input is connected to a shaft 24. The shaft 24 is the output shaft of a servo amplifier 25, which, with a frequency standard 26, comprises means for detecting the system frequency deviation from the standard or desired frequency. The output of the frequency standard 26, which is a signal having frequency fs, is supplied to one input of the servo amplifier 25, and the other input is supplied from the system power line having frequency fa.

The frequency standard 26 may be of any conventional well-known type, such as an oscillator whose frequency is controlled by a tuning fork or a crystal, and its principal requirements are that it provide an output signal of suflicient amplitude to drive the servo amplifier 25 and of a frequency which is constant to the degree required by public utility power generation systems.

The servo amplifier 25 may be one of several known designs. One servo amplifier, which is known to be suitable for this use, is described in copending application Serial No. 395,119, filed November 30, 1953, by I. J.

Larew and C. E. James, and assigned to the same assignee as the present invention. The schematic diagram of that amplifier is shown in Figs. 4 and 4a and will be described hereafter. However, the present invention is not limited to the use of any particular device and any servo amplifier that functions in the described manner may be employed. Briefly, the servo amplifier 25 compares the frequencies or phases of the two input signals and causes its output shaft 24 to rotate at a speed proportional to the frequency or phase difference between the two signals and in a direction determined by the polarity of the difference.

As was previously mentioned, the signal supplied to the servo amplifier 25 from the system power line has the actual system frequency fa, and the signal supplied thereto from the frequency standard 26 has the standard frequency fs. Therefore, when the two input signals are compared in the servo amplifier and the shaft 24 caused to rotate at a speed proportional to the frequency difference, the shaft rotates at a speed proportional to Af. The tie line load controller 2G and the servo amplifier 25 are so arranged that the differential 23 adds together the Af and kaw signals when the actual tie line load from the local to remote areas is greater than the scheduled load and the actual system frequency f., is higher than the standard frequency fs.

As was previously mentioned, system frequency and actual tie line load from the local to remote areas both increase or both decrease when the load changes in the local area, whose generation is being controlled by the system. Therefore, in that case, the differential 23 adds together the two deviation signals to produce rotation of its output shaft as a speed proportional to (M+/ruw) and in a direction determined by the polarity of the added signals. On the other hand, when the load changes in the remote area, the system frequency and tie line load vary in opposite directions, and Af and kAw tend to cancel each other and cause no change in generation in the local area. Since this effect is desired, the constant k, by which Aw is multiplied, is adjusted by varying the gain of the tie line load controller 20 so that Af and kdw just cancel each other, when the changes in system frequency and tie line load are due entirely to a load change in the remote area. The proper value of k may be found easily by experimentation. Gf course,

a constant k may be applied to the frequency deviation e signal Af rather than to Aw, so long as its value is such as to cause the two deviation signals to cancel each other when they are caused by load changes in a remote area.

It is pointed out that various means may be employed io cause rotation of a shaft, such as the output shaft G of`ditferential 23, at a speed proportional to the alge# braic sum of the two deviation signals, and the control system of the invention is not limited to any particular methods or apparatus for producing the deviation signals.

The output shaft of the differential 23,- which is rotating at a speed which is proportional to (Af-l-kAw), is connected to the rotor of a differential selsyn 28.

The construction, characteristics, and method of operation of differential selsyns are well known in the art, and need not be described in detail. It is believed snieient to point out that, when the stator winding of a differential selsyn is energized by a three-phase voltage, the frequency of the three-phase voltage induced in the rotor winding is equal to the frequency of the voltage on the stator plus or minus the speed of rotation of the rotor. For example, if the stator winding is energized by a {5S-cycle per second voltage and the rotor is turned at a speed of five cycles per second, the voltage induced I in the rotor will have a frequency of either cycles per second or cycles per second, depending on the direction of rotation of the rotor. In the present case, the stator winding of the selsyn 28 is energized from the system power line having frequency fa, and its rotor is rotated at a speed proportional to the signal (M+/cnw). Therefore, the output of the rotor of the differential selsyn 23 has a frequency equal to fa plus the signal (Af-l-kuw).

The rotor winding of the differential selsyn 28 on which the control signal appears is connected to the stator windings of a control transformer selsyn 29 whose rotor is mechanically connected to a reversible motor 30. The selsyn 29 and the motor 30 form a penalty factor unit 3l, one such unit being provided at the central control station for each controlled generating station or alternatively, for each controlled generator. For purposes of explanation, it will be assumed that a penalty factor unit is provided for each station. When motor 36 is energized by means to be hereafter described and the rotor of control transformer selsyn 29 rotated, the frequency of the station control signal induced in the rotor windings is increased or decreased from the frequency of the signal which energizes the stator winding of the selsyn. This is done in order to reapportion the load between the various controled stations of the local network.

As is well known to those skilled in the power transmission art, denite losses occur in transmitting power from a generating station to a load. In order for optimum economic system operation to occur, it is necessary to evaluate those transmission losses so that generation may be properly allocated among the various stations comprising the network. For example, it is obviously uneconornical to transmit equal amounts of power to a load from generating stations located at unequal distances from the load, assuming that both stations can produce power at the same cost. Therefore, it is necessary to reduce the power output of the distant station and increase the output of the closer station, so that the costs of power delivered to the load from the two stations are equal.

The calculation of penalty factors for the various stations of a power network is well known to those skilled in the art, and is described in an article entitled Evaluation of Methods of Coordinating Incremental Fuel Costs and Incremental Transmission Losses, byl

is'the incremental fuel cost of station n in dollars per megawatt-hour, and

Lt P1l is the incremental transmission loss of station n in dollars per megawatt-hour.

A more convenient alternative form of Equation l is (ZF l *an (2) Now lnL-L (3) (1 L where Ln is the penalty factor of steam-driven generating station n. Therefore, Equation 2 becomes:

4 LZPn/ Ln It is also known that the incremental transmission loss of a given station n may be expressed as:

where Pm are generating station or other source loadings (suc'has tie lines), and Bmn are loss formula coel'licients ldetermined as described in the aforementioned articles. For a particular station n, Equation 5 becomes:

12:2eme+*2Bn2Pn-2BMPJ-t .2B,.P.

Substituting Equation 3 into Equation 6, the expression becomes:

(7) Therefore, in order to provide the most economic operation of the generating system, itis necessary to adjust the generation of each generating station so that the incremental costs of delivered power as defined by Equation 4 are -the same for all stations. In order to dothis, itis necessary to determine for each station n 1" the transmission loss factor as defined by Equation 7 and the incremental fuel cost dan dp,

It is .then possible to adjust the outputs of the stations for equalV incremental costs of delivered power.

Referring again `to Fig. 1, in the embodiment of the invention there shown, the quantity for each of the generating stations is obtained from a transmission loss factor computer 32 to which are providedY signals proportional to the generator plant loadings, to the tie line loadings, and any loads that do not nonfarm the assumptions of Equation 5, as pointed out in the aforementioned references. Those signals may be supp-tied to the computer from conventional telemetering equipment or the like. Equation 7 takes account of all source loadings, Whether they be steam-driven generating stations, tie lines, or 'hydro-electric generating stations.

The determination of the quantity as indicated by Equation 7 may be conveniently performed by means ol' an analog computer. An instrument such as the Reeves electronic analog computer (REAC) manufactured by the Reeves instrument Cotporation, New York. to operate satisfactorily as the transmission loss factor computer 32. Alternatively,'the analog computer manufactured by Goodyear Aircraft Corporation, Akron, Ohio, will operate satisfactorily in this application.

Having obtained the quantity for each of the controlled generating plants, itis now necessary to obtain the incremental fuel cost dFn dPn for each of the plants before the incremental cost of delivered power can be determined. To obtain the incremental fuel cost, use is made of function generators 33, there being one such function generator for each controlled generating plant.

The inputs of the function generators 33 are direct,

current electrical signals proportional to the loadings, or power outputs, of each of the generating stations. These signals may be supplied from the telemeter equipments that provide the same signal Yto the transmission loss fact-or computer 32. The output of each function generator 33 is a direct voltage proportional to the incremental fuel cost dPn corresponding to the power output of the generating station at that particular time. Function generators of the type suitable for this application are well known in the art and numerous such devices are described in chapter 6 of a book entitled Electronic-Analog Computers, by Korn and Korn, McGraw-Hill Book `Company (i952). A function generator of the type that is suitable for the present application is also shown in Fig. 5 and will be later described in detail.

Having now obtained the incremental fuel cost and thc transmission loss factor (the reciprocal of the penalty factor) for each plant, it is now necessary only to divide the former by the latter to obtain the incremental cost of delivered power as dened by Equation 4. This function may be performed by a dividerV 34, there being one divider for each controlled generating station in the net work. Apparatus for performing a division of two electrical quantities is well known in the art, and examples of such devices are given in the book Electronic Analog Computers, previously referenced. The output of each divider 36 is an electrical quantity proportional to the v York, has been found in .practueincremental cost of delivered power for each of the controlled generating stations.

ln order to'vary the generation of each controlled generating station to cause generation with equal incremental cost of delivered power, use is made of an averaging amplifier 35 and a plurality of balancing amplifiers 36 that work in conjunction with the penalty factor units 31, previously described. The output of each divider 34, which is a quantity proportional to the incremental cost of delivered power of each generating station, is supplied to the input of the averaging amplifier 3S whose function is to provide an output sign-al proportional to the average incremental cost of deliverd power. Such amplifiers are well known in the art and are available commercially, as well as being described in the above-mentioned book, Electronic Analog Computers.

The outputL of the averaging amplifier 35, which is an electrical signal proportional to the average incremental cost of delivered power, is supplied to each of the balancing amplifiers 36. Of course, the number of balancing amplifiers corresponds to the number of controlled generating stations in the system. The signal proportional to the incremental cost of delivered power for each controlled generating station is also supplied to a corresponding balancing amplifier, which serves to compare that signal with the signal supplied from the averaging amplifier and provide a direct curent output signal whose magnitude is proportional to the difference between the two input signals and whose polarity is determined by the polarity of that difference. Although the invention is not limited to the use of any particular balancing amplifier, one such device that will perform satisfactorily is illustrated in Fig. 8 and will be hereafter described. The output of each balancing amplifier 36 is supplied to the motor 3l) forming part of the penalty factor unit 31 for each controlled station.

If the incremental cost of deliverd power for a par ticular station is less than the average cost of delivered power, the output of the balancing amplifier 36 for that particular station would be such as to cause the motor 30 in the penalty factor 31 to rotate in a direction to vary the control signal supplied through the selsyn 29 in a direction to call for an increase in the generation of that particular station. Similarly, if the incremental cost of delivered power of a particular station is greater than the average incremental cost of delivered power, the selsyn 29 in the penalty factor unit 31 for that station would be rotated by the motor 30 in a direction to call for a decrease in generation of that station.

In general, the various stations of the power system will be at various points from the load center and, hence, will have different penalty factors. Therefore, it is usually necessary to provide a penalty factor unit 31, a function generator 33, a divider 34, and a balancing amplifier 36 for each controlled station in the system. It

is understood that, although in the diagram of Fig. 1 only three such sets of equipment are shown, the number of them in a system would correspond to the number of controlled generating stations which, of course, is not limited to any particular number.

it is pointed out that the economic generation control features of the invention have been described with reference to steam or thermal powered generating sta"- tions. However, those features may be adapted to controlling the generation of a combined thermal and hydroelectric system as described in an article entitled Short Range Economic Operation of a Combined Thermal and Hydroelectric Power System, by Chandler, Dandeno, Glimn, and Kirchmayer, published in AIEE Transactions, vol. 72, part Ill, 195.3.

The control signal for each station may be supplied from the rotor of control transformer selsyn 29 in each penalty factor 31 to a frequency divider 37 to reduce'the control signal frequency to make it more suitable for transmission. T he control signal may then be transmitted lll tothe individual station by a carrier current transmitter 38 or other conventional means such as yleased wire, micro wave transmission, etc. For example, transmission means that are suitable for the present use are described in U. S. Patent 2,701,329, issued February l, 1955, to E. E. Lynch and G. S. Lunge, and assigned to the same assignee as the present invention.

It is noted at this point that the entire system requires only one primary detector to detect frequency deviation and one primary detector to detect tie line load deviation, both being located at the central control station. However, the central control station will generally contain a penalty factor unit 31 and its associated energizing equipment, and control signal transmission means for each controlled station in the network.

Referring now to the embodiment of the control signal responsive means located at each generating station, as shown in Fig. 2, the control signal sent to each station is received by suitable means such as a carrier current re ceiver 40, and its original frequency is restored by a frequency multiplier 41. Thus, the output of frequency multiplier 41 is once again a control signal having a frequency fa-l- (Af-i-knw). Of course, the control signal may have been modified by the penalty factor unit 31 (Fig. 1), but it is assumed for clarity of explanation that it has not been modified. It is noted that Fig. 2 illustrates only one station, and the control elements for only two generators within that station are shown. Itis understood, however, that the equipment shown will be duplicated in each controlled generating station of the network, and that each station may comprise any number of individual generators, each with the specific pieces of control equipment which serve them.

The control signal from frequency multiplier 41 is supplied to one input of a servo amplifier 42, which is similar to the servo amplifier 25 previously described, and may be identical, if desired. The second input to the servo amplifier 42 is from the system power line having a frequency fa. Because of the similarity of the servo amplifiers 25 and 42, it is sufficient here to note that the servo amplifier 42, compares the frequency of the control signal Afa-l- (Af-i-knw) with the system frequency fa, and produces rotation of an output shaft 43 at a speed proportional to the frequency or phase difference between the two signals and in a direction determined by the polarity of the difference.

The shaft 43 is connected to the input shaft of a gear box 44, which has output shafts 45, 46, 47, 48, and 50. The gear box 44 is of conventional design and it may be arranged to have the gear ratios between its input shaft and its output shafts of any desired values. 1n the present instance, it has been found satisfactory to have ratios between the input shaft and output shafts 45, 46, 47, 48, `and 50 of 1:1, 4:3, 2:1, 4:1, and 10:1, respectively, so that output shaft 45 rotates at the same speed as.

the input shaft and the other output shafts rotate at fractions of this speed. The output shafts 45, 46, 47, 48, and 50 are connected to the rotors of differential selsyn 51, 52, 53, 54, and 55, respectively, and the three-phase stator windings of these selsyns are energized from the system power line having frequency fa. Therefore, the frequency of the voltage induced on the three-phase rotor winding of differential selsyn 51 will be 32,-1-(nf-l-kaw),v

the frequency of the voltage induced on the rotor of selsyn 52 will be fa plus 75% of the area requirement signal (Af-l-knw), the rotor of selsyn 53 will have an induced voltage of frequency f, plus 50% of the area requirement signal, the induced voltage on the rotor of selsyn 54 will be of frequency fa plus 25% of the area requirement signal, and the frequency of the induced voltage on the rotor of selsyn 32 will be fa plus 10% of the area requirement signal.

The induced voltages on the rotors of the differential selsyns are each coupled to a contact of a station controlrate selector switch 56, the rotors of selsyns 51, 52,

Beaconr 53, 54, and S5 being connected to contacts 57, 58, 60, 6l, and 62, respectively, and the system power line having frequency fa is connected to contact 63. It is pointed out that the rotors of the differential selsyns which are connected to selector switch 56 -have three-phase windings thereon, and, therefore, selector switch 56 would normally have three banks of contacts, but is shown as a single-bank switch for purposes of explanation. A station control rate selector switch 56 is provided for each of the generators comprising the controlled station, so that the control rate may be individually selected for each generator in accordance with its response capabilities. This is necessary becasue generating equipments of various ages and physical conditions have different rates at which they are capable of varying their output, In the drawing, only two such control rate selector switches are shown connected lin parallel, but it is understood that any number may be employed which corresponds with the number of generators under control within the station.

A movable contact selector arm 64 of each unit control rate selec-tor switch'56 is connected to the stator winding of a control transformer selsyn 65, which is thus energized by one of the three-phase voltage outputs of the differential selsyns Slt-55. The'control transformer the control signal supplied from the control rate selectorV switch 56 to be modified in accordance with the economic considerations for each of the generator units.

At the generating station, a voltage proportional to the incremental cost of generated power at the station is provided from a potentiometer 68 connected across a standard voltage source 70, and having a movable contact pick-off ami connected to an output shaft 71 Yof the gear box 44. The potentiometer 68 serves to integrate the output of the servo amplifier 42, and the voltage present on the potentiometer pick-off arm represents the incremental cost of power generated by that station. A Voltmeter 72 connected to indicate the voltage on the arm of potentiometer 63 may be calibrated in terms of station incremental cost.

T he incremental cost voltage appearing on the movable contact arm of potentiometer 68 is supplied to parallelconnected potentiometers 73, one of which may be required for each generator within the station. Each potentiometer 73 ser-ves to multiply the incremental cost voltage by a fuel cost factor, and produce on its movable contact pick-off arm a voltage which is proportional to Y the incremental heat rate for the corresponding generator.

If the fuel cost is the same for all of the generators with- Y in the station, only one potentiometer need be used,

and all of the outputs taken from the contact pick-offv arm of the single potentiometer. However, it is possible that the fuel cost will vary from generator tol generator, and the settings of the potentiometer pick-off arms will vary slightly from one to another to take this dierence into account.

The incremental heat rate signal from the pick-off arm of each potentiometer 68 is supplied to the input of va function generator 74. Although the present invention is not limited to the use of any particular function generator, one which is known to be suitable is similar in construction to the function generators 33 (Fig. 1) and will be described later with reference to the diagrams shown in Figs. 5-7.

Each function generator 74 serves to produce an output signal that is proportional to the power output of the generator corresponding to the incremental heat rate signal supplied to the function generator. signal of each function generator 74A is supplied toan .The outputY input of va balancing amplifier 75 which will be described hereafter. A second input of each balancing amplier 75 receives a signal from a telemeter receiver or other conventional means, which is proportional to the output of the controlled generator. Each balancing amplifier then compares the amplitudes of the two input signals, and, if there is a difference in their amplitudes, energizes the motor 66 in the load proportioning unit 67 to modify the control signal output of the control transformer selsyn 65 to cause the output of each generator to be varied until the signal proportional to that output is the same as the output signal of the corresponding function generator 74. Thus, each generator tends to adjust its output to the point of most economic loading.

if the generators are being run by turbines, it may be desirable to adjust the function generators 7 4 to produce an output signal that is proportional to the valve position of the turbine, rather than the power output of the generator. In this case, the signal connected to the second input of the balancing amplifier will be proportional to the valve position of the turbine, rather than to the actual power output of the generator.

The control signal output of the rotor of each control transformer selsyn 65 is supplied to one input of an electronic positioning unit 76. The electronic positioning unit 76 serves to compare the phases (or frequencies) of two input signals and, if they do not agree, to provide an output which will actuate other apparatus to correct the difference. The present invention is not limited tothe use of any particular electronic positioning unit, but one such device which is known to be suitable is described in U. S. Patent 2,796,556 granted June 18, 1957, on copending application Serial No. 395,118, tiled November 30, 1953, by J. J. Larew, and assigned to the same assignee as the present application; the schematic diagram of that device is shown in Fig. 9 of the present application and will be described hereafter.

As previously mentioned, one of the inputs to the electronic positioning unit 76 is from the rotor of control transformer selsyn 65, which signal has a frequency equal to fa-i-(AH-knw), if it is assumed for case of explanation that the control signal has not been modified by the penalty factor unit or the load proportioning The second input to the electronic positioning unit is from the rotor of a control transformer selsyn 77, whose threephase stator winding is energized by the system line power having lfrequency fa. From the previous description of the operation of a control transformer selsyn, it is apparent that the frequency of the voltage induced in the single-phase rotor winding of selsyn 77 will be the same as that which energizes the three-phase stator winding, unless the rotor is rotated to add to or subtract from that frequency. The electronic positioning unit 7o serves to compare the frequency of the signal connected thereto from the rotor of control transformer selsyn 65 in the load proportioning unit 67 with the frequency of the signal connected `thereto from the rotor of control transformer selsyn 77. If the frequencies of the two signals differ, the electronic positioning unit 76 energizes a reversible synchronizing motor 78, which changes the speed-level setting of a governor S0 of the generator prime mover (not shown). The synchronizing motor 73 is reversible, and. the operation of the electronic positioning unit 76 is such that it can cause the synchronizing motor to turn in either direction, depending on the direction of frequency difference between the two input signals to the unit. The synchronizing motor 78, which is connected to the governor 80, is also connected through a reduction gear assembly til to the rotor of control transformer selsyn 7 7 Thus, as the Y synchronizing motor 78 rotates in response to an output from the electronic positioning unit 76, it also rotates the rotor-of control transformer selsyn 77, and thus changes the frequency of the signal induced on the rotor. There fore, the rotor of selsyn 77 will be so rotated that the frequencies of the signals supplied to the two inputs ofthe assassin 13 electronic positioning unit 76 agree. Thus, the speed-level setting of the governor 80 is changed by an amount proportional to the integral with respect to time of the area requirement signal (Af-l-kAw).

Consider now the operation of the tie line load-frequency control system shown in Figs. l and 2 in response to load changes in the local area, which is controlled by the system, and in the remote area to which the local area is joined by one or more tie lines. For purposes of eX- planation, it will be assumed thatl the change in load occurs at either the local area or the remote area, and the complex situation in which load changes occur simultaneously in both areas will not be considered. It is pointed out, however, that in the more complicated situation the operation of the control system is basically the same as in the simple situation involving only one load change, and if the control system is installed in both the local and remote areas, the system will be quickly restored to normal, even in the case where the loads at the local and remote areas change simultaneously in varying proportions.

First, consider the case in which the load in the local area decreases, thus resulting in an excess of generation in this area. It is apparent that in this case the system frequency will increase, and the power transmitted over the tie line from the local area to the remote area will also increase. Thus, the signal w, from the telemeter receiver proportional to actual tie line load supplied to the tie line load controller 2i) will be larger than the signal ws, which is produced within the tie line load controller and is proportional to the prescheduled value of the line load. The output shaft 22 of the tie line load controller will rotate at a speed kow proportional to the tie line load deviation Aw=wsw. Therefore,because the actual tie line load from the local to remote area is greater than the prescheduled value, output shaft 22 will rotate in a direction to cause the control signals developed on the rotor of selsyn 2S to call for a reduction in system generation.

The actual system frequency fa is greater than the standard frequency fs and, therefore, when the standard frequency fs is compared with the system frequency fa by the servo amplifier 25, the output shaft 24 will rotate at a speed proportional to Af and in a direction to produce control signals that call for a reduction in generation. Thus, the two deviation signals add together to produce the area requirement signal which is also proportional to change in incremental cost of generated power for the system.

The area requirement signal (Af-pkoi@ is then combined with a signal having the system frequency fa by the selsyn 28. In this particular case, the diiferential 23 causes the rotor of the selsyn 28 to rotate in a direction to subtract the area requirement frequency from the system frequency in order to reduce lthe power generated within the controlled area. The frequency of the signal supplied from the rotor winding of the selsyn 23 to the stator windings of the control transformer selsyns 29 in the penalty factor units 31 is equal to fa-(Af-l-kdw).

Assuming that the penalty factors for the stations are not changed and the motors 3i) in the penalty factor units are not energized, the control signals pass through the penalty factor units with their frequencies unchanged and are then transmitted to each generating station through a vfrequency divider 36 and transmitter 37 or other conventional means.

At each individual generating station (Fig. 2), the control signal is returned to its proper frequency by the frequency multiplier il and is fed into the servo amplifier d2, which serves to subtract the system frequency fa from the signal and cause the output shaft 43 to rotate at a speed proportional to the area requirement signal (Af-I-knw). Various proportions of this signal are again combined with a signal having the system frequency fa by the diierential selsyns Slt-55. The unit control rate having frequency 1d selector switches 56 permit the selection of any one of variously proportioned control signals (or only the system frequency, from contacts 63) to be sent to the control signal responsive means for the individual generators within the station.

For purposes of illustration, it is assumed that the 100% control signal is selected from the unit control rate selector switches 56 and supplied to the stator windings of control transformer selsyns 65 in the load proportioning units 67. As the shaft 43 rotates, it positions the movable contact pick-oif arm of the potentiometer 68 from which is derived a voltage proportional to the incremental cost of delivered power for the station. This incremental cost voltage is modified by the fuel cost factor of the various units by potentiometers 73, and the resultant incremental heat rate voltage is fed to the function generators 74. The function generators 74 serve 'to produce an output voltage which is proportional to the output of each 'generator corresponding to the incremental heat rate voltage sent to the function generators, and the balancing amplitiers 75 serve to compare the actual outputs of each generator unit with the desired outputs as derived from the function generators. lf these two factors are not the same, the balancing amplifiers 75 produce voltages to cause rotation of the motors 66 in the load proportioning units to add to or subtract from the frequency of the station control signal to attain the desired output from the individual generators. A generator control signal is supplied from the rotor of each control transformer selsyn 65 to one of the inputs of each electronic positioning unit 76. This signal, which in the present illustration has a frequency fa-(AH-kAw), assuming that no economic factor modification has occurred, is compared in phase or frequency with the systern frequency fa; and, because there is a difference in phase or frequency between the signals, synchronizing motor 78 is energized to reset downwardly the speedlevel setting of governor 80. Simultaneously, the rotor of control transformer selsyn 77 is rotated to send back to the electronic positioning unit a signal whose frequency is the same as that provided the other input of the electronic positioning unit from control transformer selsyn 65. Thus, the motor 78 rotates until the control signal frequency f}-(Af-1-klw) equals the system frequency fa, and the speed-level setting of governor 30 remains at its new level.

It is again pointed out that this same action occurs for each of the controlled generators within each station; and, because each generator unit is provided with a unit control rate selector switch 56 and with an economic control portion comprising a function generator 74 and a balancing amplifier 75, the speed-level settings of the governors for the various generators may differ one from the other. Similarly, because in the central control station there is a penalty factor unit 31 for each controlled station of the network, the control signals received by the station may differ from each other.

As a further example of the operation of the system, consider the case in which the load in the local area increases, thus causing a decrease in the frequency of the system and a decrease in the tie line load from the local area to the remote area. In this case, the operation of the various components of the system would be like that previously described for the case where the load in the local area decreased, except that the direction of rotation of the various motors and selsyns would be opposite to that previously described. For example, the frequency deviation o7 and the tie line load deviation Aw would both be positive. Thus, the rotation of the output shaft of differential 23 in the system master controller would be in an opposite direction from that previously described, and the frequency of the control signal derived from the selsyn 28 in the master controller would be greater than the frequency fa of the system. At each individual controlled station, the electronic positioning l units 76 would cause the synchronizing motors 78 of the generators to turn in a direction so as to increase the speed-level settings and increase the generation, rather than decreasing it as was the case in the first example considered.

A different situation arises when the load in the remote area changes. For example, if the load in the remote area decreases, the system frequency fa tends to increase and the power flow over the tie line from the local area to the remote area decreases. Therefore, the frequency fa is greater than the standard frequency fs, and the frequency deviation Af is negative. However, the actual tic line load wa decreases, thus becoming less than the scheduled load ws, and the load deviation Aw is positive. Therefore, kAw and Af tend to cancel each other; and the .output shaft of differential 23 will not rotate. Therefore, the signal sent from the control station to each individual generating station will be a signal having frequency fa, Vin the absence of a change of penalty factor. At the individual stations, this frequency is compared with the system frequency fa; and, because these two frequencies are the same, there will be no control action exercised in the individual stations. Therefore, the outputs of the generators in the local area will be unchanged by such a load change in the remote area.

As a final example, consider the situation when the load in the remote area increases, thus tending to decrease the system frequency and increase the power flow from the local to the remote area. In this case, the frequency deviation kaf is positive, and the tie line load deviation Aw is negative. Thus, it is apparent that once again the output shaft of differential 23 will not rotate; and the control signal received by each of the stations will once again have the system frequency fa, which will initiate no change in the speed-level setting of the governors of the generators.

The consideration of the last two examples brings out another of the outstanding advantages of the invention. It Was noted that, although the control signal was such as to cause no change in the speed level setting of the generator governors, nevertheless the control signal was present. Thus, the control signal may be continuously monitored to insure that it is present, even though the corrective portionv of the control signal may be zero.

ff the control system is to be used to control the generation of a local power system, which is operating independently and is Anot connected by tie line means to a remote power system, the tie line load controller and the differential 23 maybe dispensed with or made inoperative and the shaft 24 connected directly to the rotor of the differential selsyn 28. in that case, any load changes in the independent local area will cause a change in system frequency, which will cause the power system to vary its generation to return the system frequency to its standard value, and thus cause the generation output to match the system load.

in some interconnected power systems, it is customary for one system to vary its generation to maintain only the tie line load at its prescheduled value and to ignore system frequency changes. In that case, the servo amplifier 25, frequency standard 26, and differential 23 may be eliminated or made inoperative, and the output shaft 22 of the tie line load controller 20 connected directly to the rotor of the differential selsyn 23. The frequency of the control signal will be dependent only on the tie line load deviation, and the generation of the local system adjusted only to maintain the tie line load at its prescheduled value without regard for system frequency.

In the embodiment of the invention shown in Fig. l, it is necessary that the operator at the central control station know of the order in which the various generators in each of the generating stations under control are added to the line or removed from the line. This is necessary because each function generator (Fig. l) has Vset into it a curve representing the incremental fuel cost for a stall' tion which is a composite'of the incremental fuel cost curves of the generators within that station. Therefore, if the curves of the individual generators change, or if the generators are not added to and taken off the line in predetermined order, the curves set into the function generator 31 for that station will be incorrect relative to the actual curve ofthe station. Of course, as will be pointed out hereafter in the detailed description of the function generator, it is possible to vary the curves set into the device without undue diiculty.

The necessity of informing the control station operator before adding or removing generators may be obviated by providing a control signal for each generator under control rather than for each generating station as in the embodiment just described. This means, of course, that in the control station (Fig. 1) a penalty factor unit 31, function generator 33, divider 34, balancing amplifier 36, and control signal transmission means mustrbe provided for each' generator under control. The station equipment (Fig. 2) may be greatly simplified, however. it is only necessary to provide for each generator control signal receiving means, a servo amplier 42, selsyn Si, positioning unit 76, selsyn 77 and gears 8i. The function generator 74 and balancing amplifier 75 for each generator are eliminated because their function is performed by the function generator 33 and associated equipments at the central control station. With this arrangement, it is not necessary that generators be added to or removed from the line in predetermined order.

It is apparent that control of the individual generators is independent of the characteristics of their synchronizing motors. This is so because there is feedback from each synchronizing motor 78 through the control transformer 77 to its electronic positioning unit 76, so that each motor is individually controlled.

It is pointed out that the economic factor control equipment, which is one of the most important features of the present invention, permits continuous, automatic adjustment of the outputs of each generating station and the generators therein to maintain the most economic operation possible.

It will be recalled that the tie line load deviation signal kAw is provided by the tie line load controller 20 in the central control station (Fig. 1). The tie line load con- 'i troller 20 receives a direct current signal wa proportional to actual tie line load, compares it to a signal ws provided within itself and proportional to desired or scheduled tie lline load, and causes rotation of the output shaft 22 at a speed proportional to the difference Aw. The present invention is Vnot limited to the use of any particular device for the tie line load controller. However, apparatus that operates satisfactorily in this application is described in U. SqPatent 2,753,505 granted on copending application Serial No. 395,117, filed November 30, 1953, by J. I. Larew and K. N. Burnett, and assigned to the assignee of the present invention. Reference is made to that application for a detailed description of the device, which will nowV be described in more general terms with reference to its schematic circuit diagram shown in Fig. 3.

The direct current input signal w,l proportional to tie line load, which is to be compared with a reference signal ws proportional to desired or scheduled tie line load, may be connected between a pair of input terminals and 9i. A smoothing filter comprising a resistor 92 and a capacitor 93 serves to shunt any alternating current ripple voltage that might be superimposed on the direct current signal. The reference direct current signal wS is derived from a potentiometer 94 connected across a pair of standard direct voltage cells 95a, 95h. The movable contact arm of the potentiometer 94 is connected in series circuit relationship with the capacitor 93. A voltmeter 96 is connected between the movable arm of potentiometer 94 and the juncture of the cells 95a, 95b; the voltrneter may oe calibrated directly in terms of desired tie line load 17 w5,and theload schedule varied by changing the position of the potentiometer movable arm.

The series circuit formed by capacitor 93 and potentiometer 94 is coupled to a means for converting direct current to alternating current. Such means comprises a solenoid operated chopping device 97 having a movable armature 98 and a pair of fixed contacts 100 and 101 connected to opposite ends of the center tapped primary winding of a transformer 102. The center tap of the transformer primary winding is connected to the movable arm of a potentiometer 103, one end of which is connected between the cells 95a, 95b through a potentiometer 104 adapted to function as a variable resistance. The function of potentiometers 103, v104 will be explained hereafter.

The operation of a device such as the chopper 97 is well-known in the art, and will not be described in detail. Briey, however, the chopper is actuated by an operating winding 105 that is energized from a low voltage secondary winding 106 of a main power transformer 107, whose primary winding 108 may be connected to the 60-cycle power line of the system. As the operating winding 105 of chopper 97 is energized by the 60cycle alternating current owing therethrough, the armature 98 is caused to vibrate between contacts 100 and 101 at the frequency of the current through operating winding 105. Thus, during one-half of each cycle of the energizing alternating current supplied to the operating winding, current ows through one-half of the primary winding of transformer 102, and, during the other half cycle of the energizing current, current fiows through the other half of the primary winding. The direction of this current flow depends on whether the direct current signal (ws-wa) supplied by the series circuit comprised by capacitor 93 and potentiometer 94 is positive or negative; 'that is, the current flow through the primary winding of transformer 102 may be from the ends of the winding to the center tap if the deviation signal Aw is of one polarity, and from the center tap to the ends of the winding if the signal is of the other polarity. An alternating current signal is induced in the secondary winding of transformer 102 which is in phase with the voltage appearing across operating winding 105 if the direct current signal Aw supplied to armature 98 is of one polarity, and which is 180 out of phase with the voltage across operating Winding 105 if the direct current signal Aw is of the opposite polarity. Because the phase of the voltage across operating winding 105 (which is derived from secondary winding 106 of transformer 107) is taken to be in phase with the 60-cycle line voltage, the 60-cycle line voltage which energizes the primary winding 108 of transformer 107 may be used as an alternating current reference voltage. Hence, by comparing the phase of 'the alternating current signal induced across the secondary winding of transformer 102 with the phase of the 60-cycle line Voltage supplied through primary winding 108 of transformer 107, an indication may be obtained of the polarity of the direct current signal Aw supplied by capacitor 93 and potentiometer 94.

The alternating current signal induced in the secondary winding of transformer 102 is coupled to the control grid of the first stage of a conventional three-stage resistance-capacitance coupled amplifier, comprising triode electron discharge devices 110, 111 and 112, along with lthe necessary anode and cathode resistors and the interstage resistor-capacitor coupling circuits. Voltage is supplied to the anodes of the discharge devices 110, 111 and 112 by a half-wave rectifier 113. Rectifier 113 is energized from a secondary winding 114 on main power transformer 107, and the direct current output thereof is supplied through a conventional resistance-capacitance filter network 115 to the respective anode electrodes of electron discharge devices 110, 111 and 112.

The amplified alternating current signal appearing on 18 the anode of triode electron discharge device 112, which is of opposite phase from that appearing across the secondary winding of transformer 102, is vsupplied by means of a coupling capacitor 116 and grid biasing resistor 117 to the control grids of a pair of duo-triode electron discharge devices 118 and 120. The anodes of duo-triode 118 are connected together and to one end of a secondary winding 121 of power transformer 107, and the anodes of duo-triode'120 are connected together and to the other end of secondary winding 121. The four cathodes of the duo-triode tubes 118 and 120 are interconnected and have a common cathode resistor 122. From the foregoing description, it is apparent that the anodes of duo-triode 118 are energized by a potential that is 180 out of phase with the potential that energizes the anodes of duo-triode 120. A center tap on secondary winding 121 of transformer 107 is connected to a field winding 123 of a two-phase motor 124, whose other winding 125 is connected through a capacitor 126 to the 60-cycle line voltage input, previously mentioned as being the phase reference alternating current voltage for the device. The capacitor 126 in series with winding 125 serves to insure that the current through that winding is out of phase with the current through winding 123 for reasons which will be pointed out later. A capacitor 127 connected across winding 123 serves as a iilter and to tune the output circuit of the final stage.

As was previously mentioned, an alternating current signal is provided on the input of the three-stage amplifier which is in phase, or 180 out of phase, with the 60- cycle line voltage input. Therefore, the alternating current signal supplied to control grids of duo-triodes 118 and is also in phase, or 180 out of phase, with the 60-cycle line voltage which energizes primary winding 108 of transformer 107. Hence, the alternating current signal is either in phase with the anode voltage of duotriode 118 and 180 out of phase with the anode voltage of duo-triode 120, or vice Versa. If there is an alternating current signal on the grids of the duo-triodes 118 and 120, one of the duo-triodes will conduct more heavily tlian the other, and there will be more current through motor winding 123 during one half-cycle than during the other half-cycle to cause the motor 124 to rotate in one direction or the other. Of course, the direction in which the motor turns is determined by the polarity of the direct current deviation signal Aw, and its speed of rotation is related to the amplitude of the deviation signal. It is pointed out that at balance, when there is no vdeviation signal Aw, equal currents ow through motor winding 123 during both halves of each cycle, and the motor 124 does not rotate.

The motor 124 is connected to a feedback signal developing means comprising a direct current tachometer 128 which when rotated in one direction produces a direct current feedback voltage of a certain polarity, and when rotated in the opposite direction produces a direct current feedback voltage of the opposite polarity, the amplitude of the voltage being proportional to the speed of rotation. The direct current Voltage produced by tachometer 128 as motor 124 rotates is fed back to the deviation signal producing circuit, previously described, lto linearize the output speed of motor 124 with respect to the direct current deviation signal input. One side of the output cir-cuit of tachometer 128 is connected to the juncture of cells 95a, 951;, and the other side is connected to the movable contact of a three position selector switch 130. One of the fixed contacts of the three position selector switch is connected directly through a resistor capacitor network 131 to the previously unconnected end of potentiometer 103, whose other end is connected to potentiometer 104, the two potentiometers functioning as a voltage divider. The remaining two fixed contacts of selector switch are connected to the network 131 through a pair of batteries 132 and 133, arranged with opposite polarities. By proper selection of the xed contacts of selector switch 130, a positive or negative voltage maybe added in series with the direct current feedback voltage developed by tachometer 123. Because the movable contact of potentiometer 103 is co-nnected to the center of the primary winding of transformer 102, the feedback signal is added to the direct current deviation signal developed by capacitor 93 and potentiometer 94.

It is apparent that as various amounts of the direct current feedback signal from tachometer 123 are fed back to the deviation signal developing circuit comprising potentiometer 94 and capacitor 93, the slope of the curve of the output motor speed versus input volts will vary, and will be stabilized thereby. It is also apparent that if the direct current feedback voltage is too small, this output curve will cease to be linear. It is the function of the voltage divider comprising potentiometers 103 and 104 to set the amount of direct current feedback voltage available. Thus, the setting of the movable contact of potentiometer 104 is determined by the amount of D. C. feedback voltage required to linearize the output curve of motor speed versus deviation signal Aw, and is influenced Vby the slope desired for this curve. The desired slope for the curve of output speed versus input deviation signal may be obtained by adjustment of the movable contact arm of potentiometer 103.

For purposes of explanation and description, let it be assumed that during the iirst half of each cycle of the 60-cycle reference voltage, current iiows through the upper half (as seen in the drawing) of the primary winding of transformer 102, and during the second half of each cycle, current flows through the lower half of the winding. Let it also be assumed that the direct current tie line load signal wa connected between input terminals 90 and 91 is more positive than the reference signal ws taken from potentiometer 94, and, therefore, that current will liow downwardly through the upper half of the primary winding of transformer 102 during the lirst half of each cycle, and upwardly through the lower half of the winding during the second half of each cycle. Thus, it is assumed that the alternating current signal appearing across the secondary winding of transformer 102 is in phase with the 60- cycle voltage supplied across the primary winding 103 of power transformer 107. Therefore, because the arnplier comprising discharge devices 110, 111 and 112 has an odd number of stages, the signal appearing on the anode of dis-charge device 112 is 180 out of phase with the reference 60-cycle supply voltage. Now, let it also be assumed that the voltage on the anodes of duotriode 118 is in phase with the 60-cycle power line reference voltage, and thus voltage on the anodes of duotriode 120 is 180 out of phase with this reference voltage. Therefore, if during the rst half of each cycle the alternating current signal appearing on the control grids of the duo-triodes is negative, while the voltage appearing on the anodes of duo-triode 118 is positive and the voltage on the anodes of duo-triode 120 is negative, no current will flow through either of the duo-triodes if the alternating current signal is great enough. How ever, during the second half of each cycle, the alternating current signal placed on the control grids is positive, the anodes of duo-triode 118 are negative, but anodes of duotriode 120 are positive. Therefore, current will flow through duo-triode 120 during the second half of each cycle. Thus, depending on the connections, the phase of the current flowing through motor winding 123 is either 90 ahead or 90 behind the phase of the current flowingthrough motor winding 12.5, and the motor will turn in a certain direction. As the motor 124 turns, tachometer 128 generates a direct current feedback voltage part of which appears across potentiometers 103 and 104. That feedback voltage is of proper polarity to act as a negative feedback, in the manner ,well-known in electronic circuitry. Thus, the direct current deviation signal tends to be reduced as the motor speed is increased 'zo and the net effect isV to cause the speed of the motor 124 to stabilize at a speed kAw which is proportional to the direct current tie line' load deviation signal Aw.

Conversely, if the direct current tie line load signal w, connected across terminals 90 Iand 91 is negative with respect to the reference voltage set on potentiometer 94, during the first half of each cycle current flows upwardly through the upper half of the primary winding of transformer 102, and flows downwardly through the lower half of the winding during the second half of each cycle. Thus, the signal appearing across the secondary winding of the transformer is 180 out of phase with O-cycle line voltage which energizes primary winding 103 ofthe power transformer 107. Again, the alternating current signal appearing on the anode of discharge device 112 is 180 out of phase with the input, xand thus is in phase with the 60-cycle reference voltage. During the rst half lof each cycle, as was previously pointed out, the anodes of duo-triode 118 are positive, and the anodes of duo-triode 120 are negative. Thus, during the first half of each cycle, in the present illustration where the direct current deviation signal is negative Vlimited to the use of this particular device.

rather than positive, the control grids of duo-triode are positive at the same time that the anodes thereof are positive. Therefore, current ows through duotriode 118 during the first half of each cycle. During the second half of each cycle, the signal appearing on control grids is negative, and, therefore, if the signal is of sufficient magnitude, neither of the duo-triodes in the final stage conduct. Thus, it is seen from the two examples just described, that when the direct current deviation signal Aw is positive, current ows through motor winding 123 during the second half of each cycle, and when the direct current deviation signal is negative current ows through the motor winding during the first half of each cycle. Because of the capacitor 126 in series with motor winding 125, the current owing through motor winding 123 is either displaced in phase 90 ahead or 90 lagging the current through winding 125, and the motor turns in opposite directions when the direct current deviation signal is positive or negative. A suitable direct current feedback signal will then be developed and fed back to the input of the mechanism in the manner described above.

It is noted that the mechanical connection between the motor 124 and the tachometer 12S may be extended to provide the connection 22 between the tie line load controller 20 and the selsyn 21, as shown in Fig. l.

Figs. 4 and 4a illustrate a possible form of the scia/o amplifiers 25 and 42 previously mentioned with reference to Figs. l and 2, although the present invention is not The servo amplifier illustrated forms the subject matter of copending application Serial No. 395,119 as previously noted.

It will be recalled lthat the servo amplifiers 25 and 42 serve to compare the frequencies or phases of two input electrical signals and produce rotation of an output shaft at a speed proportional to the frequency difference.

Referring to Fig. 4, it is seen that the servo amplilier comprises a discriminator 135, which will be later described in detail, a conventional two-phase motor 136 and a conventional control transformer selsyn 137. One input to the discriminator may be from a frequency standard (as in Fig. l) or other source (not shown), and the second input is from the rotor of the selsyn 137. The stator winding of the selsyn 137 is energized by the signal whose frequency or phase is to be compared to that of the signal from the frequency standard or other source, which is connected to the first input of the dis criminator. If there is a difference in frequency or phase between the input signals to the servo amplifier, there is an output from the discriminator 135 that causes the motor 136 to rotate. The rotor of the selsyn 137 is connected to the output Vshaft'of the motor and rotates with it; as does a shaft 138 that is the output shaft of the servo amplifier.

As is well known, the voltage induced in the rotor winding of a control transformer selsyn has the same frequency as the voltage which energizes the stator winding of the transformer, if the rotor is not rotating. However, if the rotor is turning, the voltage induced across the rotor winding will differ from the stator voltage by an amount determined by the speed of rotation of the rotor. For example, if the stator is energized by a voltage having a frequency of 60 cycles per second, and the rotor is rotated at five revolutions per second, the voltage induced in the rotor winding will have a frequency of either 55 or 65 cycles per second, depending on the direction of rotation of the rotor. Thus, in the present instance, as the motor 136 turns in response to a phase or frequency difference between the servo amplifieninput signals, it will vary the frequency or phase of the second input signal to the discriminator which is taken from the rotor of the control transformer selsyn 137. It is :apparent that, when the elements are connected as a servo amplifier in the manner shown in Fig. 4, the motor 136 will be caused to rotate the rotor of selsyn 137 and the output shaft 138 at a speed which is proportional to the phase or frequency difference between the first input signal to the discriminator 135 and the signal which energizes the stator winding of the .selsyn 137.

Fig. 4a illustrates a possible form of the'discriminator 135 that is embodied in the servo amplifier shown in Fig. 4. Referring now to Fig. 4a, one of the two input signals which must 'be approximately a sine wave and mustnot be grounded, may be connected to input terminals 140 yand 141, and the second input signal connected to termina-ls 142 and 143 if the signal is not grounded, or to input terminals 142 and 144 if the signal is grounded. The input signal connected to terminals 140 and 141, which is taken to be the reference signal and will be hereafter'referred to as signal A, produces a voltage drop across resistors 145, 146, and 147, connected in series between the input terminals. Resistors 146 and 147 are of the same value, so that the signals appearing across them are equal, and resistor 145 serves merely as a series dropping resistor,

The second input signal is assumed, for purposes of explanation, to be ungrounded and connected to terminals 142 and 143. This signal, which will be hereafter referred to as signal B, appears across a resistor 148 connected between terminal 143 and the junction of resistors 146 and 147. An inductance 150 is connected lbetween input terminals 142 and the junction of resistors 146 and 147, and operates as a filter in conjunction with fa capacitor 151 connected across resistor 148, which permits the application of an alternating voltage to terminals 142 and 143 that may be of other than a strict sine wave shape.

The junction of resistors 145 and 146 is connected to the cathode of an electron discharge device 152 of the diode type, and input terminal 141 is connected to the cathode of a similar discharge device 153. It is now seen that at any instant, the signal appearing at vthe cathode of diode 152 is the vector sum of the signals appearing across resistors 146 and 148, and the signal appearing at the cathode of diode 153 is the vector sum of the signals appearing across resistors 147 and 148.

The anodes of diodes 152 and 153 are connected together through resistors 154 and 155, the juncture of which is connected to input terminal 143. Resistors 154 and 155 are of equal value so that, when diodes 152 and 153 are conducting equally, equal signals appear across the two resistors. Capacitors 156 and 157 are connected across resistors 154 and 155, respectively, and act in conjunction with the resistors to filter the output signal of the diodes. 1f the input signals to the circuit are such that equal signals appear on the cathodes of the diodes,

the diodes conduct equally, and equal negative signals will appear on the anodes thereof. If the input signals are so related that unequal signals appear on the cathodes of the diodes, as will be later explained in detail, one diode will conduct more heavily than the other, and the D. C. voltage appearing at its anode will be more negative than that appearing at the anode of the other diode.

The D. C. voltages appearing across resistors 154 and 155 also appear across resistors 158 and 160, which serve as grid resistors for two D. C. amplifiers comprising triode electron discharge devices 161 and 162. The cathodes of the triodes 161 and 162 are grounded and the anodes of the triodes 161, 162 are connected through resistors 163 and 164, respectively, to a source of D. C. voltage, which will be later described. The control grid of triode 161 receives the negative D. C. signal appearing at the anode of diode 152, and the control grid of triode 162 receives the negative D. C. signal appearing at the anode of diode 153. A capacitor 165 is connected between the juncture of resistors 154 and 155 and the juncture of resistors 158 and 166 and serves to permit the grid-tocathode voltages of triodes 161 and 162 to be zero at the balance condition. That is, capacitor 165 causes only the difference in the voltages appearing across resistors 154 and 155 to be applied to the D. C. amplifiers, and maintains an alternating current ground at the juncture of resistors 154 and 155.

The output of the D. C. amplifiers is taken from the anodes of triodes 161 and 162 and is coupled through antihunt networks to the input of a power output stage. The signal appearing at the anode of triode 161 is connected through an anti-hunt network 166 and through a current limiting resistor 167 to the control grids of a pair of triode electron discharge devices 168 and 170, which are connected in parallel and have a grid resistor 171 across which the signal appears. Similarly, the signal appearing at the anode of triode 162 is connected through an anti-hunt network 172 and through a current limiting resistor 173 to the parallel-connected control grids of triode electron discharge devices 174 and 175, where the signal appears across a grid resistor 176. The cathodes of triodes 168, 170, 174 and 175 are connected through a biasing resistor 177 to the juncture of grid resistors 171 and 176. A capacitor 180 is connected between the juncture of resistors 171 and 176 and ground, and functions in a manner similar to that of the capacitor 165 previously described.

The parallel-connected anodes of triodes 168 and 170 are connected to one end of a secondary winding 181 of a transformer 132, and the parallel-connected anodes of triodes 174 and 175 are connected to the other end of the secondary winding 181. Thus, the voltage supplied to the anodes of triodes 168, is 186 out of phase with the voltage supplied to the anodes of triodes 174, 175, and the triodes cannot all conduct simultaneously. The primary winding 183 of transformer 182 is connected across power input terminals 184 and 185, to which 110 volt 60 cycle alternating current may be connected from the system lines. A filament winding 186 on transformer 182 provides voltages of the proper magnitude for a filament 192 of triodes 168 and 174, and for a filament 193 of triodes 170 and 175.

One phase of a two-phase motor (such as the motor 136 shown in Fig. 4), which it is desired t0 have respond to a phase difference between the input signals, may be connected across output terminals 188 and 190 in the anode-cathode circuit of the output triodes. Terminal 188 is connected to the cathodes of triodes 168, 170, 174 and through biasing resistor 177, and terminal 190-s connected to a center tap of the secondary winding 181 of transformer 182. A jack 191 may be connected across terminals 188 and 190 so that the output may be monitored, if desired.

The anode voltage for the D. C. amplifier triodes 161 and 162 is provided by a power supply comprising a

Non-Patent Citations
Reference
1 *None
Referenced by
Citing PatentFiling datePublication dateApplicantTitle
US2888575 *May 7, 1958May 26, 1959Westinghouse Electric CorpAutomatic dispatching system
US3270209 *Dec 17, 1962Aug 30, 1966Leeds & Northrup CoEconomic loading of power systems
US3493733 *Dec 23, 1966Feb 3, 1970Matevosyan Paruir AvetisovichComputing device to simulate the characteristics of relative increments of specific fuel consumption
US6473744 *Aug 6, 1999Oct 29, 2002Mhc Investment CompanyApparatus and method for trading electric energy
US7064513Sep 30, 2004Jun 20, 2006J. L. Behmer CorporationPhase angle control for synchronous machine control
US8201000 *Jun 12, 2012International Business Machines CorporationComputing load management method and system
US8205106 *Feb 24, 2009Jun 19, 2012International Business Machines CorporationEnergy load management method and system
US8327170 *Dec 4, 2012International Business Machines CorporationComputing load management method and system
US8341442 *Mar 6, 2012Dec 25, 2012International Business Machines CorporationEnergy load management method and system
US20050073279 *Sep 30, 2004Apr 7, 2005J.L. Behmer CorporationPhase angle control for synchronous machine control
US20120166834 *Jun 28, 2012International Business Machines CorporationComputing load management method and system
US20120166835 *Jun 28, 2012International Business Machines CorporationEnergy load management method and system
Classifications
U.S. Classification307/57, 700/295, 700/297, 290/4.00A, 700/78
International ClassificationH02J3/04, H02J3/06
Cooperative ClassificationH02J3/06
European ClassificationH02J3/06